This week we’re going to use AI for two situations that show up again and again once you’re past 40: **researching a major purchase** and **writing a professional decline email**. One helps you avoid expensive mistakes; the other helps you protect your time and relationships without feeling guilty or awkward.

Try at least one before Sunday. That is the whole assignment.

Use Case 1 of 2

Use Case 1: Researching a major purchase decision

Let’s start with Jack.

Jack is 62, semi-retired, and lives in a 1970s ranch house on the edge of town. It’s Tuesday evening, just after dinner. He’s at the kitchen table with his laptop open, staring at an article about home backup generators. Last summer, two storms knocked out power for a full day each time. This year, his wife has started saying, “We really should do something about this.”

Jack has three browser tabs open: one from a big-box store, one from a local electrician, and one review site that looks helpful until he notices it’s covered in ads and “Top Pick” badges that don’t tell him much. He’s wondering:

- Do we really need a whole-house generator, or would a smaller one be fine? - Are these brands actually different, or just different labels? - How much should installation cost, realistically?

Underneath those questions is the real one: *I don’t want to make a $7,000 mistake.*

Jack doesn’t need more random articles. He needs a clear comparison that fits *his* situation: his house, his budget, his tolerance for risk. This is where AI can act like a calm, competent research assistant - if you give it the right details.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> I am considering a major purchase and I want you to act as my neutral research assistant, not a salesperson. > > Here is my situation: > - Age: 62 > - Location: small town in [your state or region], with 2 - 3 multi-hour power outages per year > - Home: 3-bedroom single-story house, about 2,000 square feet, built in the 1970s > - Household: 2 adults, one with medical equipment that needs power (CPAP machine, small oxygen concentrator) > - Budget: ideally under $8,000 total, including installation > > I am deciding whether to buy a whole-house backup generator. > > Step 1: Briefly list 3 - 5 realistic options for someone in my situation (including “do nothing” and “smaller portable option”), with one sentence on who each option is best for. > > Step 2: For the two most sensible options for me, create a clear comparison that covers: > - Total cost range (equipment + installation, in my region) > - Ongoing costs (maintenance, fuel, etc.) > - Reliability and risk during outages > - Impact on resale value of the home > - Convenience (noise, where it sits, how it’s used) > > Step 3: Ask me 5 - 7 clarifying questions to help narrow down what really matters most to me (e.g., budget vs. convenience vs. medical needs vs. resale value). > > Step 4: Based on my answers, give me: > - A short “If this is you, then choose X” recommendation > - A simple checklist of what to ask local installers or vendors (no brand hype, just practical questions) > > Important: > - Do not assume I’m handy or enjoy doing my own maintenance. > - Do not push a specific brand; focus on types of solutions and trade-offs. > - Be concrete and specific about numbers, but flag anything that might vary a lot by region.

Why this prompt works:

It tells the AI **who you are**, **where you live**, and **what constraints you’re under**, so it doesn’t give generic “Top 10 generators” fluff. By breaking the task into steps, you force the AI to first map options, then compare, then ask you questions, and finally recommend - very similar to how a good consultant would work. You also explicitly tell it to be neutral and trade-off focused, which keeps it from sounding like marketing copy.

AI will not know your local electrician’s prices or the quirks of your town’s permits. It will give **reasonable ranges**, but you still need to sanity-check those against real quotes. Treat this as a decision *framework*, not the final word. And be careful with “medical equipment” questions - AI can outline risks, but it cannot replace medical advice or guarantee safety.

Use Case 2 of 2

Use Case 2: Writing a professional decline email

Now meet Carla.

Carla is 54 and has run her own consulting business for eleven years. It’s Wednesday morning, 9:30 AM. She’s just finished a Zoom call with a former colleague who wants her to “just take a quick look” at a project proposal and maybe “join as an advisor for a few months.” It’s unpaid, vaguely defined, and would easily swallow twenty hours a month if she lets it.

Carla knows she should say no. Her plate is already full with paying clients, and she promised herself she’d leave room this year for travel and downtime. But this is a person she likes. She doesn’t want to sound harsh or ungrateful. She also doesn’t want to leave the door open for endless “maybe later” follow-ups.

So she does what many experienced professionals do: she leaves the email draft blank and goes to make coffee, hoping the right words will come to her. An hour later, nothing. Just a knot of obligation.

This is exactly the sort of careful, emotionally loaded writing that AI can help with - *if* you give it enough context and clear boundaries.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> I need help drafting a professional email where I **decline** an opportunity while maintaining a good relationship and clear boundaries. > > My situation: > - I am a [your profession, e.g., “independent marketing consultant”] in my [50s]. > - The person is: [brief description, e.g., “former colleague, now leading a nonprofit initiative”]. > - What they asked for: [describe in 2 - 3 sentences, including if it’s paid/unpaid, long-term/short-term]. > - Why I am saying no: [be honest in 2 - 3 sentences - capacity, misalignment, need for paid work, etc.]. > - Tone I want: warm, appreciative, clear, firm, and short (no more than 250 words). > > Please: > 1. Draft 2 different versions of the email: > - Version A: more formal, suitable for someone I don’t know well. > - Version B: more conversational, suitable for someone I like and respect. > > 2. Include: > - A brief appreciation for the invitation. > - A clear, direct sentence that says I am not able to participate. > - (Optional) One sentence that suggests an alternative (e.g., “I can recommend X,” or “You might look for Y type of person”), but only if it feels natural. > - A closing line that leaves the relationship intact but doesn’t invite ongoing negotiation. > > 3. Before giving the final drafts, list 3 - 5 “key phrases” that people in their 40s - 60s often find helpful in these emails (e.g., “Given my current commitments…”), and then weave a few into the drafts. > > Important: > - Do NOT apologize excessively. One short “sorry I can’t this time” is enough. > - Do NOT suggest I might be available later unless I explicitly say I would be. > - Keep everything in plain language - no corporate jargon.

Why this prompt works:

You give the AI **the relationship dynamics**, **your reasons**, and **the tone**, so it doesn’t produce a stiff corporate memo or a breezy text message. Asking for two versions lets you choose what fits best, and forcing it to include a clear declining sentence prevents those vague “I’m not sure” responses that people read as “try again in a month.” The instructions about apologies and availability help the AI reinforce your boundaries instead of undermining them.

AI doesn’t know the full history between you and this person. If there are sensitive past events, longstanding favors, or cultural nuances, you’ll need to adjust the wording yourself. Also, be careful with the “optional alternative” sentence: if you let AI suggest things you don’t truly want to offer (like introductions or free advice), delete those lines before sending. The goal is clarity, not creating new obligations.